This will tell you the charging capacity of your current cable+charger combo. In other words, a 100W charger + 100w cable will return "Wattage (W): 100". A 100W charger + 15W cable will return "Wattage (W): 15". A 15W charger + 100W cable will return "Wattage (W): 15".
So, you kinda need a charger that meets or exceeds the power delivery of the cable you're testing. And I believe this relies on the cable's self-identified capacity via the e-chip or whatever, so it could be fraudulent if you have some dodgy cable from an unknown manufacturer? But it's better than nothing.
There is a lot of info in `/usr/sbin/system_profile`, not sure if there's other bits related to identifying USB-C cable capacity. I'm sure there has to be something.
You can get this information from the "System Information" app in macOS as well. (At least, that's what it's called in Ventura...)
It should be under Hardware -> Power, in the section titled "AC Charger Information".
You should also be able to see connected USB-C cable capacity under Hardware -> USB. For example, mine says the following:
USB 3.1 Bus:
Host Controller Driver: AppleT8112USBXHCI
USB3.0 Hub:
Product ID: 0x0813
Vendor ID: 0x2109 (VIA Labs, Inc.)
Version: 90.11
Speed: Up to 5 Gb/s
Manufacturer: VIA Labs, Inc.
Location ID: 0x01200000 / 2
Current Available (mA): 900
Current Required (mA): 0
Extra Operating Current (mA): 0
USB2.0 Hub:
Product ID: 0x2813
Vendor ID: 0x2109 (VIA Labs, Inc.)
Version: 90.11
Speed: Up to 480 Mb/s
Manufacturer: VIA Labs, Inc.
Location ID: 0x01100000 / 1
Current Available (mA): 500
Current Required (mA): 0
Extra Operating Current (mA): 0
Any way to do similar for DP/HDMI? I've thrown away multiple cables cause I couldn't tell if the cable was just old and didn't support what I needed vs somehow broken. Looking at system report I only see `Graphics/Displays` which show's monitor capabilities but not cable. (I'm not actually sure if DP/HDMI advertise their capabilities in the same manner as USB 3).
Much like GP I've taken to labeling newly bought cables with a label maker to avoid wasting them in the future due to being unknown.
DP and HDMI cables are mostly passive with no identifying chip, so the only way to determine their bandwidth electronically is to try to bring up the link at that speed.
yeah I've seen that, would be awesome to have access to it but not worth the cost to me, was less interested in stress testing does this 100% meet spec and more interested in "this labels itself as HDMI v{X} with optional features {y}, {z}".
I mostly care as my current monitor requires HDMI 2.1/Display Port 1.4 and going through my random cables drawer was extremely tedious. As I had way to many cables.
There should be no such thing as a 15 watt cable, by the way.
Every cable is a 60 watt cable by default. 20 volts, 3 amps.
Cables designed for 5 amps have a marker chip in them. If they follow the old spec they support 20 volts, and if they follow the new spec they support 48 volts. There's not much incentive to fake these because you still have to put a chip into it and 60 watts is fine most of the time.
That sounds like you're hitting the limit from non-charging ports. Normal A to C should allow much more. But either way that's a special case here, because the limits are entirely on the USB A end. That's not a proper C cable.
An A-to-C cable is limited by what an USB-A host can supply, since that's the limit which a compliant C-device sensing an A-to-C cable or adapter would observe.
Depending on the combination of host and device, that would be 4.5W for standard USB 3, 2.5W for USB 2, up to 7.5W for USB Battery Charging (BC), or possibly more for non-USB-C PD (highly unlikely, since it would require legacy PD signalling on the voltage pins, which almost no USB-C devices support as far as I know).
I strongly suspect it's USB BC, rounded up to 8W by Apple.
As an aside: Apple's USB power controllers used in their laptops are surprisingly flexible based on my observations: In addition to the obvious USB-PD, they support at least USB BC and Apple's proprietary USB-A resistor-based protocol. I was able to successfully charge a MacBook Pro at 12W and even 5W supplied by an old iPad/iPhone charger and the appropriate combination of adapters once.
Apparently, USB-C cables are supposed to contain an “e-marker” chip that can tell you its bandwidth, wattage, and other specs. It would be nice if this information showed up somewhere when you plugged in an empty cable into your Mac.
I’ve also tried to get that information on macOS before, but annoyingly all promising (i.e. related to the SOP protocol which is how the computer talks to the chip in the cable) log lines seem to be redacted.
This is doubly frustrating in that it shows that it’s clearly available to the OS (and not handled by a lower layer controller), but also leaves the faint hope that Apple might implement just such a feature in the future.
My suspicion is that this is security related. I'm on the lookout for a brand-name quality Thunderbolt-capable cable with an LCD showing current power.
> So, you kinda need a charger that meets or exceeds the power delivery of the cable you're testing. And I believe this relies on the cable's self-identified capacity via the e-chip or whatever, so it could be fraudulent? But it's better than nothing.
These can be useful, but ultimately a watt meter tells you what the device is actually using at the moment, which is not only limited by the cable and source capabilities, but also by the power sink device's current needs.
Looks like PD spec varies voltage all the way up to +48V which supports 240W of power, so 5A@48V.
Using a calculator like at https://www.omnicalculator.com/physics/wire-size a 1m cable with a relatively user friendly max wire temp of 40c asks for a diameter of 0.6mm. Looks like there is 4 VBUS pins and even tho it's reversible I imagine for full power it requires all 4 (or all 4 must be connected to "support" the spec).
In theory you could have reduced wire diameter and spread current across the 4 VBUS wires, or have a single wire of the correct diameter and then connect to all 4 VBUS at the connector but everything is stranded wire in those cables anyway and I assume there's some complex maths to do to figure out the best for it.
It is super freaking cool, though. It makes me wonder if many household appliances can't just use usb c. I don't know how much further they can raise the voltage beyond that though seeing as there's all the electrical safety rules on conductor spacing etc to avoid arcing/flash over.
The whole cable is smaller than 5mm diameter. Less rubber jacket, and shield, it might just be 3mm inside. And there are ~18 lines in there? I expect the power lines are larger than the differential signalling lines, but I'm just realizing how much I'm trusting this vendor.
That's much more than big enough. 1mm diameter was the overkill number, and 3mm diameter is nine times as big while you only need two of them. Even accounting for insulation and the smaller data wires, that's plenty.
I've been using a nice Brother labeler for years to label all my power plugs. eg this power plug goes to this router etc. since a lot of them are subtly different.
I once mixed up two different netgear router power plugs - a new version and an older version. I had the newer netgear router hooked up to the old plug - which gave slightly less amperage - and everything worked fine until it would randomly die. It took me a long time to figure out that it was dying when it was under load and trying to pull more power than was available.
I've also started doing this, with USB-C to C cables you might want to get something like the C2C caberQU cable tester as a lot of cheap type c cables don't actually wire all the conductors.
Also I like to note the cable length on the label, measured by myself since manufacturers always count the connectors on the total cable length and on some of them that can add up to almost 10cm
As a techie who understands that there are many different types of USB-C and HDMI cables depending on what you need to use it for, it's an incredible amount of effort to find the right thing to buy.
When you consider that most brick and mortar stores (I'm looking at you, Currys PC World) massively rip you off with £100+ gold plated HDMI cables, and the search results on Amazon are filled with knock-off Nigel rubbish (or worse, listings that out-right lie to you)... it's a total minefield!
Imagine what it's like for the average consumer! A complete disaster!
Especially tedious on Amazon when you have a seller list the same cable in various lengths. The reviews are all commingled, so you'll get plenty of valid five star reviews for the short cables while the long ones don't support the bandwidth.
And who knows how many reviews are from non-techies who are just getting a picture on the screen and leave a five star review without knowing / caring if they actually got HDR or not.
This would be difficult with HDMI cables – unlike USB-C, they’re not electronically marked.
Only HDMI 2.1 introduced link training (before that it was just the source picking a resolution that the sink supports and hoping for the best!), but even that is an end-to-end thing; the cable is not part of the conversation, so you wouldn’t know if the cable is bad or the socket/internal wiring beyond the cable on either side.
Oh, sorry, I thought this was in the sub thread about “why not mark cables in software”!
Fully agreed – the maximum supported data rate would be even more important to be printed on the cable for HDMI.
One reason I could see why manufacturers would be hesitant to do this is that some future improvements might not have higher requirements for cabling, so the printed data rate might be underselling actual capabilities.
I personally prefer the aesthetics of the white Amazon Basics [0] and/or Infinite Cables [1] hdmi 2.1b [2] 48gbps cables, but as long as they're certified and pass the totalphase cable test I guess it doesn't really matter
I hooked up a new monitor and had the hardest time trying to figure out why things weren't working. Finally swapped the HDMI cable and it worked perfectly. Gah.
And where is the list of cables that have passed the compliance test and use the cable logo?
I can't say I've ever seen one of the package logo's, and yes I've been aware of them for a while now.
This is obviously a case of "mistakes were made" but I think the fact that they haven't mandated the logos as part of the license to use the USB specifications says a lot about the companies that run the organization. (or maybe their products).
Edit: There is this, but no actual type-c -> type-c cables, https://www.usb.org/products. Oh maybe they are under "retail -> cable assembly" but I still can't find one with a logo, which is the same problem. Once I buy it, how do I know what it can do in a year or two when I pull it out of a pile of cables.
Or just remember what the U is USB actually stands for.
We've been through the mutually incompatible plug standards. USB was introduced to solve that. Now we're back to mutually incompatible cable standards.
>Now we're back to mutually incompatible cable standards.
Not exactly, we have a single standard with different tier cables for power and speed.
If you buy a USB-C cable with high power delivery, it will work for lower powered devices as well (but the inverse will not). If you buy a USB-C with high speed, it will work for devices will very low throughput rates too.
So buy the most expensive highest power+speed USB-C version, and you don't have to worry, it would support all (or close) lower-level uses.
At this point I'd probably pay ~$100 / year for a subscription service that annually sends me 4-5 cables of the "max rating." I don't want to waste my life keeping on top of the every-changing spec, but would 100% just turn-over my cables once a year to ensure highest performance.
Perhaps change their color every year (red -> green -> blue -> purple) to keep track of which are the latest cables.
I have dozens of almost identical black USB-C cables :|
Just get certified Thunderbolt 4 and be done with it then. Cheaper is TB3. If you stick with the legit ones you're done and it will basically support everything.
That's my solution: I just buy Caldigit TB4 cables and I know they will do everything required.
There's also the Dockcase Smart USB-C Hub 10-in-1 Explorer which exposes what's actually happening across your various connections including USB-C. Unfortunately it doesn't seem to support video passthrough over USB-C, so it isn't perfect.
Certified Thunderbolt cables with a length over 4 feet don't really exist, though. If you need a cable longer than that which can handle 100w+ and video/data, you're going to have a much harder time finding something decent.
In the end, the technology just doesn't exist to send 100w and full data speed over the usb-c interface longer than that reliably (and economically enough for the market) so it just isn't an option to go longer. You'll have to find a different solution (optical plus separate power) or do a cross-your-fingers method with a non-certified cable that may or may not work under different circumstances.
I don't really need 40gbps Thunderbolt, but a 10ft cable that can do 100w, usb3 10/20gbps, and alt modes would be perfect for me. I have sadly not been able to find a cable that does these things reliably. Thunderbolt is just a way to certify these lesser features plus a PCIe alt mode.
Alongside this amazing subscription service, I would love a service that provides a list of tech brands that never do any kind of false advertising, so that I know I can buy from them with confidence. Recently had a bad experience with a spate of SSD external enclosures that all were under-spec, and worked with Ex-Fat but not APFS+ (which requires 100% USB spec compliance to work properly). Ended up finding a brand called "Cable Matters" which was a bit more expensive but was up to spec and well worth the extra cost.
Thanks for that!
Always hard to find the right tools, with a recommendation it is much easier.
(Sometimes, you don't know if it's just you? Or there really are quality differences.)
Exactly what I was thinking yesterday. Why not continue with different colors on the USB plugs, as they sort of did with blue for usb c or was it USB 3? Or. Ah. Can't keep up.3.1? (Some motherboards tried, but there weren't cables in same color.)
You mean USB 3.2 Gen 2x1 aka USB 3.1 Gen 1 aka USB 3.0 aka Super Speed USB?
Or do you mean Thunderbolt 3 which supports USB 3.2 Gen 2×1 aka USB 3.1 Gen 2 which was called SuperSpeed+ USB but is now SuperSpeed USB 10Gbps?
Or the newest USB4 Gen 4, USB4 Gen 3×2, USB4 Gen 3×1, USB4 Gen 2×2 and USB4 Gen 2×1 which supports Thunderbolt 3?
Maybe they could start with the upcoming Thunderbolt 5 which follows the USB4 2.0 spec, not to be confused with USB4 Gen 2(x1/x2).
I'm not sure we have enough colors to enumerate the different type of USBs. Going by the naming schemes it seems like there's already not enough numbers. /s
Most of those are market noise because of manufacturers wanting to sell things which aren't under conformance.
Thunderbolt 4 and 5 are external conformance marks, because the USB-IF can't be motivated to stop the abuse their own members are causing.
For USB-C to C cabling, there is only bandwidth and power capacity. The operational modes supported by the devices you are connecting with that cable is out of your control (except purchasing replacement hardware).
All USB 3 devices are compatible with USB 3.2. They roll up the requirements for everything in there in addition to compatibility across modes. The 3.2 spec does not say you should "now call it USB 3.2 Gen 2x2".
The only actual name changes I can think of
- SuperSpeed+ became Superspeed 10 Gpbs.
- The branding improved around USB4 to combine power and bandwidth capabilities into a combination of name and graphical mark for cable labelling and the like
All the "Gen" bullshit was explicitly considered to be an internal designation. You can blame tech journalism for bringing that out into the world, because the USB IF never promoted that as a marketing term.
I just watched a puff piece on Adam tested comparing apple thunderbolt $150 cables to introductory $5-10 usb cables. They seemed more in love with the tech to visualize the cables than actuallu giving meaningful information and like for like comparisons.
Not eMarkers but retimers / redrivers. Basically, once you go beyond a certain length you need the cable to "boost" the signal. The tested cable is longer than that, so it needs active electronics.
Data channel: no is a charging-only cable, *preferred* for dealing with untrusted sources of power. While it might violate spec it has a very valid use case.
So you're saying the data pins inside the device are attached to something, and when that something makes a connection it stops the device from drawing power?
Does this happen when you plug them into chargers and when you plug them into hubs?
This happens with some devices. It would have been better if the standard had made that failure mode impossible, but it's not the data lines causing the problem. They failed to put a resistor in the device that connects one of the sense pins.
Are you saying that every USB-C cable is required to support DP alt mode? I’ve heard that’s not the case — that active cables, for example, will usually not support DP functionality.
better than microUSB that's fair, but far worse than lightning. Engagement is terrible, it collects debris and it's inconsistent (often too tight or too lose).
This has been done for years on HDMI/Cat cables, but instead of on the connector on the cable itself. Why make the connector larger to print very little detail on it, if you have literally a whole meter to write complete sentences.
This is roughly the same solution, but in my opinion more poorly implemented. So please make more USB cables like HDMI or Cat cables, print all the relevant info on the cable.
You jest, but flexibility and endurance is important for consumer cables, and except for LAN parties, Ethernet doesn't get moved much (and is much cheaper to replace).
But what's your solution for braided cables? Because I'll pick braided over "just plastic" any day of the week, and there is no way you're going to cleanly print information on the braid.
Braiding is not fabric: fabric is a tight-weave, stable sheet, where the fibers are locked in place because it's been loomed specifically for that purpose. That's basically the opposite of what braiding is meant to do: each braid is explicitly meant to move independently, and constantly, so even if you managed to print something on the braining, in a way that looked fine coming right out of the printer, whatever you printed will be unrecognizable after even a few days of coiling and uncoiling the cable, let alone the years of use that apply to USB cables.
I do agree a braid allows more movement and flexibility than fabric. However, I can't imagine the fibers really move that much to the point whatever gets printed will become entirely illegible. It is not like the fibers in the jacket are going to permanently move several millimeters in its normal straight shape.
They don't have to, every fiber just needs to move 1/10th of a millimeter in completely different directions for text to turn into complete nonsense. As for "it's not like"... that's literally why we use braided cables, that's exactly what they do near the connector over time =)
(or any other place that bends frequently, so for cables that get coiled and tossed into a bag and then uncoiled and used a lot, that's basically "the entire cable")
By move I mean permanently in a new position. I agree while it's being flexed the text is going to get distorted, but flexing it back would mostly move those fibers back. If they always kept sliding forward, you'd pull all the fibers out after flexing it a few times. They're mostly going back to where they were when you flex the cable back.
I've written on braided ropes before with markers and they're still very legible despite being used a lot. Make a mark on a shoelace and flex it a bit. Does the line become completely messed up?
I don't know, is this really much more ridiculous than e.g. the various Ethernet standards (I can never remember which one the "normal" Gigabit Ethernet is), mobile data protocols (remember HSDPA, CDMA, 1xRTT and all that?) etc.?
And to the USB-IFs credit, they did eventually come around with the much saner bandwidth-focused names. As far as I know, even "SuperSpeed" and "USB 4" are now gone as speed designators; now it's just "USB xGbps", e.g. USB 5GBps for the case of what used to be "USB 3.2 Gen 1x1".
The good thing here is that they seem to be converging on names that contain the speed itself in plain text. "USB4 40Gbps" is as helpful as a name can get.
It's certainly much less bad now than it used to be, but the existence of "x1" connections (half lane count and bandwidth as its contemporary "x2" sibling) does undermine trust. Is a cable advertised with the number "20" really an x2 cable accurately described according to "Gen 2" metrics, or is it really just an x1 cable, daringly called "20" because that's what a gen 3x1 connection could achieve?
Looking at the Wikipedia page I get the impression that the new naming simply ignores the x1 links.
That's like an invitation for misleading product descriptions. Maybe it would have been helpful to assign slightly off but clearly confusion-resistent numbers to the x1 siblings? Perhaps taking a page from supermarket pricing, "4.99Gbps", "9.99Gbps", "19.99Gbps"?
All USB-C cables have 4 high speed wire pairs (or 0). There's no such thing as a USB-C cable that limits you to x1. x1 or x2 is entirely up to the devices attached.
So if a USB-C cable is advertising "20Gbps", you want them to be talking about x1, because that means it's a better cable.
(And if it's not USB-C on both ends, then it's always x1. It's never ambiguous.)
> Looking at the Wikipedia page I get the impression that the new naming simply ignores the x1 links.
Yes, and that probably makes some sense: All compliant USB-C cables (other than USB 2 only) need to have two high-speed lane pairs, i.e. four high-speed lanes in total. (I'm not sure why that's the case, i.e. whether the aim was to avoid further confusion or if it's required for orientation-reversible operation in some ports).
Whether the host/device actually support lane bonding or not is arguably not really the cable's choice, so it makes some sense to specify whatever the cable would support in an optimal configuration (i.e. maximum number of lanes used at the maximum supported modulation rate).
Considering the number of features and capabilities that can be given to a usb-c cable, I think a better alternative would be to take cue from the humble resistor and use color coded bands around each of the ends. Text can only really cover 1-2 capabilities and it’d be nice to have a way to know exactly what a given cable can do, even if it requires a decoder ring.
But not accessible. It's generally better to use colors AND designs (dashed, dotted, etc) over color to prevent color-blind issues, what I see as red may differ from what you see as red.
If I understand it correctly most parameters can be negotiated by the usb controller of a device. It should be possible to plug it into a device and see the specs of the cable and the charger somehow.
Yeah. A lot of IT people would pay for a box that you plug a USB cable into and it tells you what the cable can and can't do. Simply reading out the specs, not actually testing whether it can perform to spec. (Although a simple test of the latter should be possible. Baby computer, it sends data at all supposedly supported speeds and sees if it gets the results. That would catch the ones that couldn't but could miss a flaky one.)
There are only two types capabilities of a cable speed (480 Mb, 5Gb, 20Gb, 40Gb or 80Gb) and power (60W (3A) or 240W (5A)) everything else is depends on the host.
To use any of the USB-IF logos (including the official ones that clarify power and signal compatibility) a product has to go through the certification process: https://www.usb.org/logo-license
USB-IF also maintains a list of products that are certified: https://www.usb.org/products
The question then is, “Why is this still confusing?”. The answer is probably that the popular retail and e-commerce channels are flooded with products that aren’t certified, and consumers don’t realize that a certification process even exists.
The problem is that you can flood the market with 2 categories of product:
1. A cheap version that didn't bother to get certified (to save cost).
2. A fraudulent version that relies on the existence of a poorly regulated market (like Amazon) that allows "mistakes" in advertising details. That market is already flooded with cheap uncertified cables that do work, making it trivially easy to falsely advertise cables that don't work.
What needs to happen is for USB-IF to minimize #1 by making their standardization process as accessible (and affordable) as possible. Otherwise, the bulk of cables that customers are willing to buy will be uncertified.
This seems like a business opportunity. It should be straightforward (not _easy_, but straightforward) to manufacture high quality, reasonably priced cables and sell them direct. Establish a rep for consistent quality & fast shipping and carve out the 'customers who give a shit' market.
The problem is your interface with those customers.
The customers we are talking about both care about cost and are willing to put more effort into their search to find an affordable quality product.
From that customer's perspective, the best method is to look through Amazon, Ebay, Alibaba, etc. listings to find every product that claims their desired spec under a reasonable price ceiling. After that, the customer need only filter out sketchy listings. This is most reliably done by checking reviews. It's also done somewhat by checking the coherency of listing details: self-contradicting details are a common method for fraudulent sellers to filter for unscrupulous victims - the same process our scrupulous customer is performing here, played in reverse.
So what can you do as a seller to reach this customer? Make detailed listings, and get reviews. Any more work than that is, unfortunately, a wasted effort.
Word of mouth _works_, but takes some time. You want to get to the point where when non-tech folks ask their tech friends what USB cables to get, they just send them to your site because they know everything's good there.
So long as the port itself is full-featured, it's good to keep the plugs the same.
The real issue here is with our marketplaces. Amazon in particular is notorious for allowing counterfeits with very little recourse available to the end customer. This has heavily incentivized uncertified products, because there is not enough value in accurate advertising to pay for certification in the first place. Shoppers who care know to trust reviews instead of listings. Shoppers who do get a fraudulent cable will just get a refund; and the seller will just keep selling bullshit.
The marketplaces that care about accuracy love this situation. They get to take a larger cut from heavily overpriced cables, because they can brag about certification. Why would Best Buy ever try to get more competition? That would only hurt their margins.
All of this talk about testing and labeling doesn't make sense to me or likely any layperson. I don't want to have to know what capabilities a cable needs and then compare it to a label and I really don't want to have to test a cable to see if it will work for my intended use case.
I want to be able to grab any USB cable and have it work regardless of my intended use. If that is impossible because the standard is so complex then we really should have different connectors. The entire point of it being universal is that regardless of circumstance it will work. Somehow that has morphed into the connector being universal while the actual utility of each cable is not; the latter of which is the only thing people really care about.
Not being able to plug a mouse or a keyboard into a 40GBps USB4.0 laptop connector just because those devices only need the USB2.0 protocol is a dumbarse idea.
Just like having to pay for a 40GBps USB4 controller in a mouse which needs USB 1 speeds.
Despite all the whining here from "hackers", the current situation is undoubtedly better.
Terrible idea. Most of the time I just want to charge stuff, which any old USB-C mostly works fine for up to like 60 watts. And very few peripherals actually need more than like USB 2.0 speeds.
They have the same plug so we can choose anyone we want to use. It's way better than the previous "oh, no, my computer only has 1 parallel port, we can't connect this one".
Annoyingly while that’s true on Apple devices there’s a plethora of cheap laptops where only one of the USB-C ports is capable of DisplayPort, and a different one can do USB-PD. Good luck working out which is which.
If I read this correctly, those logos are for "packaging and marketing". Unless they are on the cable itself, as Elgato does, this is gonna be of little help when I'm looking for a cable in my cable box even six months later. Or if you have shared cables in a household.
Yes, see the "Cable logo" column in https://www.enablingusb.org/certification , though it's still the manufacturer's option whether to put the packaging logo on the packaging or the cable logo on the cable or both.
That does not help with counterfeits. We need an easy way to check the cable instead. Best would be if you can plug both ends into computer and the controller measures resistance (charging capability) and data bandwidth.
Or at least some standalone tester, shouldn't be hard to build , dunno about the bandwidth part.
Ideally you'd want to measure actual rather than reported charging capacity. For about $15 you can get a USB load generator and a USB current monitor on AliExpress.
For the newer USB-PD modes, you also need to know the maximum safe voltage (and related construction features, such as CC pins being a bit shorter to enable a safe/spark-free connection teardown), or you risk arcing.
AFAIK all cables are de-facto rated up to 30V, as you won't save money if you make insulation even thinner. So this is non issue in practice. Plus you can't really measure breakdown voltage without destroying the cable.
And I think one important feature is having shorter CC pins in the connector, so that the power source can shut down the high voltage to avoid arcing altogether. That seems to be a feature distinct from insulation.
By being careful? If you were otherwise going to use it without testing you're not any worse off plugging it into a test setup up to your expected load.
You can step up the load gradually and keep an eye on it.
I would be more worried about the cheap no-brand test gear catching fire. I wouldn't plug my 100W charger into it.
It wouldn’t cause more fire incidents than just using the cable without testing, right?
The tests would only run for a tiny period of time. Most fires come from too much current through too small conductors for too long. It takes time for heat to build up to fire causing levels.
The type of time required to actually try to charge something.
I would rather have a counterfeit cable that is clearly labeled. At least then I can trivially explain the fraud that was committed.
Without the labeling, every discussion about a counterfeit must include redundant clarification. Most fraudsters will take that as an opportunity to talk in circles.
Can't the host and guest measure the voltage on each end anyway? They'd just need to communicate it with each other. This would give a full end-to-end voltage drop measurement (including the plug and socket connections at both ends).
It'a weird that this isn't normal.
I bought an "AlzaPower" cable, which is just a rebrand of some chinese manufacturer by a local reseller ("czech amazon"), and they have their wattage as well as speed on them (100W/5Gbps). Then also a samsung usb c cable, which was pure white, has nothing on it and doesn't seem to support any fast charging or higher speeds..
5Gbps is conveniently unambiguous, but 10 and 20 exist in variations that use four lanes and in variations that use two lanes. According to Wikipedia, the symbols with a number are only specified for the four lanes version, but there's little doubt that half-width cables will also be be advertised with whatever bandwidth latest devices could funnel through.
That's cool and very useful, should have been in the USB specs ("connectors shall have supported version and speed imprinted").
But realistically, I can't remember the last time I cared about the data speed of a USB cable, I use USB cables almost 100% for charging. I have a few memory sticks I plug in from time to time, but those get plugged in direct without a cable.
What a mess… couldn’t this be done in software instead? As in, if you try to charge your phone with a cable that doesn’t support fast charging, the phone could detect it and pop up an alert. I’m not sure the average consumer would know what to make of these labels, and that’s assuming the cable manufacturers would comply and put accurate labels in the first place.
That's way too slow and way too late. It doesn't help finding the cable in a drawer, and it certainly doesn't give you any assurance it's a power-only cable before plugging in some unknown power supply.
But yeah, let that happen too. It's also useful.
> assuming the cable manufacturers would comply and put accurate labels in the first place
Putting the wrong label on the cable is plain simple fraud. You can easily get some civil recourse, and if you are feeling revengeful, you can make a criminal complaint against the seller. Compare that with our current situation.
That’s how it’s already done. I don’t know why no device actually exposes that information, but the e-marker in every USB-C 3.0 (or beyond) cable tells the device exactly that: The highest supported physical data rate and charging speed.
That's not true for PD charging cables though right? They are generally USB2 speeds if they have data pins beyond what's required for PD charging at all but they're just direct electrical cables without any chips as far as I know.
It would seem that with e-markers in USB-C cable (and assuming that USB 3/4 still expose device information like that to the host), an extended version of this should be possible: "Your computer and this device support x Gbps, but your cable caps out at y Gbps; use a better cable to make this connection faster".
They can do the opposite direction too, detecting a USB2.0 cable when a higher speed one is needed. It still boggles my mind that no corresponding warning pops up.
Even without an e-marker, there has to be some way to know that the client device can support a higher speed and warn the user - it seems like USB device capabilities supports that?
But it doesn't warn "charging slowly because the cable you're using doesn't support fast charging" vs "charging slowly because the charger you're using doesn't support fast charging". It tells you there's a problem, but not what the cause is even when it can determine the cause.
There you go. Seems like software could display this in a much more useful way than these labels ever could. Even the example in the article isn’t complete, it lists the data rate but doesn’t say anything about the amperage support for charging.
Just a simple information window that lists the max charging Wattage and data transfer speed of the current cable/charger/device configuration would suffice.
At least the label on the cable is useful for those of us that understand and doesn't intrude on those that don't
A popup we all have to suffer wether we want to see it or not or if we understand it or not. Oh guess I could disable it right? Yeah another task for me to do on every device I own, no thanks.
This is an excellent design solution, my only comment would be the type layout could be nicer.
The lowest power USB-C cable is still 60 W which is more than enough for any phone (aside from some Android phones with proprietary 100W charging protocols that need their own custom chargers anyway)
I've wondered why consumers don't usurp the naming conventions for USB? For whatever reason, the USB consortium completely screwed up with all the variants which led to headaches and money wasted on cables. Why not just self-organize online and apply meaningful names to devices.
So, for instance, a 10gb USB3 with oval plug and power delivery could be called "USB3P", 40gb could be called "USB4P", and so on, regardless of what advertisers or the organization calls it (don't get hung up on my example name, the community could collaborate and vote on the best nomenclature).
Tech reviewers could publish lab results of each cable SKU (and laptop model number) and provide lookup tables. Then sell cheap stickers or bands so you can label your own cables. Just completely bypass the standard and put the power in the hands of the people.
This entire mess happened because consumers usurped the naming convention. The intended marketing term is just "USB-C 5Gbps", "USB-C 20Gbps" and so on.
The "GenX" stuff was only ever intended to be an internal technical designation, but journalists who didn't bother to actually read all documents started using that term with the general public.
A lot of USB-C cables only support USB 2.0 speeds (480 Mbps = 60 MB/s). I found out this while searching for a USB-C cable for a HDD enclosure. There are a ton of "charging" cables, but very few "fast data" cables, at least in the $10-15 range. I'm sure for $100 you can find a cable with everything in.
Yes, but that's because manufacturing faster cables costs more. If all cables would have to carry USB 3+ speeds, you just wouldn't have 10$ charging cables.
It's so refreshing to encounter a manufacturer that gives a hoot about users. And then I feel sadness for realizing that this situation should be normal and routine, and it is so much not that it deserves an HN frontpage when somebody finally does the right thing.
It ultimately won't tell me if the cables are actually connected inside but I have two USB testers that can read and print the information on each emark (if it's present) in the cable, which is a pretty useful thing to have when grabbing a mystery usb-c cable. I have the avhzy ct-2 and ct-3 testers, there's a bunch of other distributors of the same thing as well. The FNB48 looks like same board, different programming.
I used to be able to recommend the avhzy ones but during the chip shortage they started using some clone chips inside which don't appear to be as accurate.
The USB standards body really should fix this by requiring all USB-C cables to indicate what their capabilities are. It seems insane to make them all the same shape and not to have a requirement for some other way they can be easily differentiated.
This is one of the two things that makes me dislike USB-C (the other is that I find the physical connectors more fragile than the predecessors).
Is there any simple way to check the capabilities of a cable, without using specialized equipment?
For me it's always a gamble, with my devices I actually can barely even see how fast they are charging on a specific cable/charger combo. Only by measuring charge time and then trying to calculate how many watts they delivered.
Personally, no. Testing equipment is the best way if you want to be sure.
If you buy directly from a reputable brand, you may be able to trust what they've said a given product can do. If you buy from a reseller like Amazon, that all goes out the window and you should assume nothing.
Worth noting the testing equipment for this isn't that expensive these days. You can pick up an FNB58 USB tester for $60.
You really shouldn't have to though. At the very least every phone on the market ought to be able to just read the e-marker chip and tell you what each cable supports. Very annoying that they don't because the information is there and they have a way to read it, it's just not exposed to the user.
I would be so happy if I could find a manufacturer that sells all common USB-C cables (to USB-C, lightning, micro-usb, mini-usb, ...) in good quality, with reliable specs, all common lengths and common color variations. A plus would be to even have the option to get one or both ends 90 degrees angled.
Personally I prefer my charging cables to be red, really helps people not to trip over them. In other places the color preferably matches the environment.
If you want a good, reliable tester, I think $10 is a bit limiting personally. You don't need to spend $100 but $10 is stretching it really thin. GIGO.
The difference is that the USB-IF logos to indicate bandwidth/charging capacity on packaging and plugs require certification, while numbers written arbitrarily on the plug don't require testing and don't have to be true.
lol you do realize manufacturers will just lie, don't you?
They'll print the usb logo or whatever they like regardless of licensing restrictions. Did you always check for the official logo on CDs and was it always standard conforming if present?
No, the consumer-facing name is “SuperSpeed USB 5Gbps”.
Gen 1x2 has been deprecated as far as I understand, so there is no more ambiguity when just referring to modes by their maximum data rate.
USB 4 would make this even more complicated, but they fortunately have decided that 3x1 (20 Gbps just like 2x2) also does not get a marketing designation, so there’s also no more ambiguity there.
"Gen 1x2" wasn't deprecated because as a marketing term it was never a thing to begin with. And they recently dropped the whole "SuperSpeed" part so now it's just "USB 5Gbps".
There was a backlash when Apple tried to put microchips and DRM into USB-C cables. I hate Apple, but I would fully support them on this.
USB-C cable problem is 10 years old. In 2013 Google introduced Pixel laptop, and some guy from Google bought all USBC cables on Amazon, only to test them and write reviews!
Microchips in cables are already a semi-optional part of the USB-C standard, mandatory for charging with more than 3A... The backlash was against rumors that Apple was planning an additional paid certification program which would mean only cables whose manufacturers paid for Apple certification would be able to achieve higher speeds...
Markers are also mandatory for all speeds above USB 2.
This part of the spec is actually quite well thought out.
It’s a shame that no device I know just displays the marker data (e.g. a message like “this connection might be limited by your cable”, assuming it’s also possible to figure out the other end’s highest theoretical speed, or simply the cable capabilities in some system menu).
"Cable chips" - or as they're actually called "eMarkers" are mandatory for all USB-C cables above basic capability.
What Apple wanted was to leech on every single cable sold for their own profit with their proprietary licensing.
You already have what you wanted - now you're just demanding Apple tax for no reason. Remember, you can always just buy that 120$ Apple cable and be a happy camper. Right now. With no extra DRM.
_Optimization/pasted_image009.png){kind=link}
I've resorted to physically labeling them with a Brother label printer.
For MacOS, here's a terminal snippet to determine the current charging capacity of a cable.
This will tell you the charging capacity of your current cable+charger combo. In other words, a 100W charger + 100w cable will return "Wattage (W): 100". A 100W charger + 15W cable will return "Wattage (W): 15". A 15W charger + 100W cable will return "Wattage (W): 15".So, you kinda need a charger that meets or exceeds the power delivery of the cable you're testing. And I believe this relies on the cable's self-identified capacity via the e-chip or whatever, so it could be fraudulent if you have some dodgy cable from an unknown manufacturer? But it's better than nothing.
There is a lot of info in `/usr/sbin/system_profile`, not sure if there's other bits related to identifying USB-C cable capacity. I'm sure there has to be something.